•  
  •  
 

Keywords

first-year students, retention, probation, predictive models

Abstract

The study used pre-college variables in the prediction of retention and probation status of first-year students in learning communities at a regional public university in South Texas. The correlational study employed multivariate analyses on data collected from the campus registrar about three consecutive cohorts (N = 4,215) of first-year students. Logistic regression models were developed to predict retention and probation status without respect to learning community membership, as well as for each learning community category.

The logistic regression model to predict retention regardless of learning community membership included five pre-college variables, while the model to predict probation status included eight pre-college variables, five of which overlapped with the retention model. The models for each learning community contained different sets of predictor variables; the most common pre-college predictors were high school percentile and the number of days since orientation. The results of the study provide practical implications for the learning communities program, as well as learning community scholars interested in targeting interventions to the students who need them most.

Share

COinS